Cognitively Healthy Nonagenarians in the Cross Cohort Collaboration (CCC) - Half of all persons who develop clinical dementia become symptomatic after age 85 years, whereas most studies of dementia have focused on younger patients in their 70s. Compared to dementia beginning at a younger age, dementia in the oldest-old has a more heterogeneous, multifactorial etiology. Although, Alzheimer disease remains important there are greater contributions from vascular brain injury, systemic disease and dysfunction that indirectly affects the brain, and recently described, poorly understood brain pathologies. Vascular and lifestyle risk factors may contribute differently to risk of dementia in the oldest-old. Conversely, it is also important to learn what resilience factors permit persons to remain alive and cognitively normal as nonagenerians. Studies that enroll persons aged 85+, often lack information from when these individuals were middle-aged. Longitudinal cohort studies that enrolled participants between ages 45 and 70 years and followed them past age 80 years, till they died or developed dementia, would be ideal study samples, but each such cohort usually has only a small number of participants surviving beyond age 80. One solution is to combine data across multiple longitudinal studies with harmonizable protocols. The Cross-Cohort Collaboration Consortium (CCC) was established in 2018, as an offshoot of the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium. The Neurology working group within the CCC (led by MPIs Seshadri, Ikram, Dufouil, Debette, Satizabal) brought together 8 cohorts (RF1 AG059421) to study late life dementia. We now request a renewal of this grant to expand and continue the collaboration. We will add 5 new cohorts reaching 13 cohorts, to study 102,285 participants, 37,803 with brain MRI, and add a new MPI in Suchy-Dicey who will bring her experience working with AD in American Indians. We propose the following aims. Aim 1: To relate various risk factors, systemic illness and MRI markers in midlife (ages 45 to 70) to (a) the risk of late life dementia, and (b) the probability of reaching age 85 (+/- 5) years, alive and dementia free (‘wellderly’). Aim 2: To assess (a) the impact of multimorbidity examining how persons with two or more chronic conditions differ in life-expectancy and dementia risk and (b) to combine various risk and resilience factors, multimorbidity and MRI measures using artificial intelligence and machine learning to create parsimonious models that predict late-onset dementia and wellderly status. Aim 3: To examine the biological pathways underlying the observed associations using pathway analyses and structural equation modeling to explore the mediating role of plasma proteins. The proteins we study will include known biomarkers of amyloid, tau and neurodegeneration (Aβ40 Aβ42, p-tau181/217, NfL, GFAP) in 37,824 individuals, as well as an array of 3000+ plasma proteins in 38,067 persons (new assays on Olink Explore 3072 + available data) so we can identify previously unsuspected biology using Mendelian randomization methods to examine causality. Aim 4: We will investigate any effect modification by sex-, age-, race-, ethnicity-, urban or rural residence, menopause, APOE genotype on findings from Aims 1-3. We hope to uncover putative drug targets to reduce risk of late life dementia.
